Quality indicators for the care and outcomes of adults with atrial fibrillation
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Europace (2020) 00, 1–17 GUIDELINES
doi:10.1093/europace/euaa253
Quality indicators for the care and outcomes of
adults with atrial fibrillation
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
Task Force for the development of quality indicators in atrial
fibrillation of the European Heart Rhythm Association (EHRA) of
the European Society of Cardiology (ESC): Developed in
collaboration with the Heart Rhythm Society (HRS), the Asia Pacific
Heart Rhythm Society (APHRS), and the Latin-American Heart
Rhythm Society (LAHRS)
Elena Arbelo (Chair)1*, Suleman Aktaa2, Andreas Bollmann3, André D’Avila4,
Inga Drossart5, Jeremy Dwight6, Mellanie True Hills7, Gerhard Hindricks3,
Fred M. Kusumoto8, Deirdre A. Lane9, Dennis H. Lau10, Maddalena Lettino11,
Gregory Y. H. Lip9, Trudie Lobban12, Hui-Nam Pak13, Tatjana Potpara14,
Luis C. Saenz15, Isabelle C. Van Gelder16, Paul Varosy17, Chris P. Gale2, and
Nikolaos Dagres3 (Co-chair)
Reviewers Serge Boveda18 (review coordinator), Thomas Deneke19, Pascal Defaye20, Giulio Conte21
Radoslaw Lenarczyk22, Rui Providencia23, Jose M. Guerra24, Yoshihide Takahashi25
Cristiano Pisani26, Santiago Nava27, Andrea Sarkozy28, Taya V. Glotzer29, Mario Martins Oliveira30
1
Arrhythmia Section, Cardiology Department, Hospital Clı́nic, Universitat de Barcelona, C. Villarroel 170, Esc 3, Planta 6, 08036 Barcelona, Spain; Institut d’Investigació August Pi
i Sunyer (IDIBAPS), Barcelona, Spain; Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; 2Leeds Institute for Data
Analytics, University of Leeds, UK; Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, UK; Department of Cardiology, Leeds Teaching Hospitals NHS
Trust, UK; 3Department of Electrophysiology, Heart Centre Leipzig at University of Leipzig, Leipzig, Germany; 4Cardiac Arrhythmia Service, Hospital SOS Cardio, Florianopolis,
SC, Brazil; Harvard-Thorndike Electrophysiology Institute, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA; 5European Society of Cardiology, Sophia
Antipolis, France; ESC Patient Forum, Sophia Antipolis, France; 6ESC Patient Forum, Sophia Antipolis, France; 7StopAfib.org, American Foundation for Women’s Health, Decatur,
TX, USA; 8Cardiology Department, Mayo Clinic Hospital, Jacksonville, FL, USA; 9Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart &
Chest Hospital, Liverpool, UK; Aalborg Thrombosis Research Unit, Department of Clinical Medicine, Aalborg University, Aalborg, Denmark; 10Centre for Heart Rhythm
Disorders, The University of Adelaide and Department of Cardiology, Royal Adelaide Hospital, Adelaide, South Australia, Australia; 11Cardiovascular Department, San Gerardo
Hospital, ASST-Monza, Monza, Italy; 12Arrhythmia Alliance/AF Association/STARS, Chipping Norton, UK; 13Yonsei University Health System, Seoul, Republic of Korea; 14School
of Medicine, University of Belgrade, Serbia; Cardiology Clinic, Clinical Centre of Serbia, Belgrade, Serbia; 15Fundación Cardio Infantil-Instituto de Cardiologı́a, Bogotá, Colombia;
16
Department of Cardiology, University of Groningen, University Medical Centre Groningen, Groningen, The Netherlands; 17Rocky Mountain Regional Veterans Affairs Medical
Center and the University of Colorado Anschutz Medical Campus, Aurora, CO, USA; 18Clinique Pasteur, Heart Rhythm Department, 31076 Toulouse, France; 19Clinic for
Interventional Electrophysiology, Heart Centre RHÖN-KLINIKUM Campus Bad Neustadt, Germany; 20CHU Grenoble Alpes, Unite de Rythmologie Service De Cardiologie,
CS10135, 38043 Grenoble Cedex 09, France; 21Cardiology Department, Cardiocentro Ticino, Lugano, Switzerland; 22First Department of Cardiology and Angiology, Silesian
Centre for Heart Disease, Curie-Sklodowskiej Str 9, 41-800 Zabrze, Poland; 23St Bartholomew’s Hospital, Barts Heart Centre, Barts Health NHS Trust, London, UK and Institute
of Health Informatics, University College of London, London, UK; 24Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Universidad Autonoma de Barcelona,
CIBERCV, Barcelona, Spain; 25Department of Advanced Arrhythmia Research, Tokyo Medical and Dental University, Tokyo, Japan; 26Heart Institute, Arrhythmia Unit, Sao Paulo,
Brazil; 27Head of Electrocardiology Department, Instituto Nacional de Cardiologia ‘Ignacio Chavez’, Mexico; 28University Hospital of Antwerp, University of Antwerp, Antwerp,
Belgium; 29Hackensack Meridian-Seton Hall School of Medicine, Rutgers New Jersey Medical School; Director of Cardiac Research, Hackensack University Medical Center,
Hackensack, USA; and 30Hospital Santa Marta, Department of Cardiology, Rua Santa Marta, 1167-024 Lisbon, Portugal
* Corresponding author. Tel: þ34 93 227 5551, Fax: þ34 93 451 3045, Email: elenaarbelo@secardiologia.es
Published on behalf of the European Society of Cardiology. All rights reserved. V
C The Author(s) 2020. For permissions, please email: journals.permissions@oup.comPage 2 of 17 E. Arbelo et al.
Aims To develop quality indicators (QIs) that may be used to evaluate the quality of care and outcomes for adults with
atrial fibrillation (AF).
...................................................................................................................................................................................................
Methods We followed the ESC methodology for QI development. This methodology involved (i) the identification of the
and results domains of AF care for the diagnosis and management of AF (by a group of experts including members of the ESC
Clinical Practice Guidelines Task Force for AF); (ii) the construction of candidate QIs (including a systematic review
of the literature); and (iii) the selection of the final set of QIs (using a modified Delphi method). Six domains of
care for the diagnosis and management of AF were identified: (i) Patient assessment (baseline and follow-up), (ii)
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
Anticoagulation therapy, (iii) Rate control strategy, (iv) Rhythm control strategy, (v) Risk factor management, and
(vi) Outcomes measures, including patient-reported outcome measures (PROMs). In total, 17 main and 17 second-
ary QIs, which covered all six domains of care for the diagnosis and management of AF, were selected. The out-
come domain included measures on the consequences and treatment of AF, as well as PROMs.
...................................................................................................................................................................................................
Conclusion This document defines six domains of AF care (patient assessment, anticoagulation, rate control, rhythm control,
risk factor management, and outcomes), and provides 17 main and 17 secondary QIs for the diagnosis and manage-
ment of AF. It is anticipated that implementation of these QIs will improve the quality of AF care.
䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏
Keywords Atrial fibrillation • Quality indicators • Outcome measures
...........................................................................................................................................................................................
Abbreviations the quality of AF care.18–21,22 However, no AF quality indicators (QIs)
have been specifically designed for the wider international community.
AF atrial fibrillation Hence, the European Heart Rhythm Association (EHRA), in collabo-
EORP EURObservational Research Programme ration with the Asian Pacific Heart Rhythm Society (APHRS), the
ESC European Society of Cardiology Heart Rhythm Society (HRS), and the Latin-American Heart Rhythm
INR international normalized ratio Society (LAHRS), established the AF QI Working Group, which was
LV left ventricle tasked with the development of QIs for the diagnosis and management
LVEF left ventricular ejection fraction of adults with AF. It is hoped that these QIs can serve as a mechanism
PRISMA Preferred Reporting Items for Systematic Review and to improve the quality of AF care, and be used by healthcare providers
Meta-Analyses to evaluate care delivery at the patient, centre, and national levels.
PROMs patient-reported outcome measures To enhance the translation of guideline recommendations into
PVs pulmonary veins clinical practice and give healthcare providers the tools to identify op-
QI quality indicator portunities for improvement, a summary of the AF QIs has been em-
QoL quality of life bedded in the 2020 ESC Clinical Practice Guidelines for AF.23 Efforts
RCT randomized controlled trial were made to ensure alignment between the developed QIs and the
ESC Guidelines for AF, which may differ from recommendations de-
veloped by other professional organizations.
Introduction
Atrial fibrillation (AF) is a key public health challenge and a major Methods
source of morbidity, mortality, and economic burden for govern- The detailed methodology for the development of QIs for the quantifica-
ments worldwide.1 Despite progress in the management of patients tion of cardiovascular care and outcomes for the ESC Clinical Practice
with AF, this arrhythmia is still a major cause of stroke, heart failure, Guidelines is published separately.24 This methodology consists of a four-
and cardiovascular morbidity and mortality globally.2 Additionally, AF step process: identification of the key domains of care; construction of
is associated with cognitive impairment,3–5 reduced quality of life candidate indicators; selection of a final QI set; and undertaking of a feasi-
(QoL),6,7 depression,8 and frequent hospital admissions.9–11 The bility assessment. In this document, we have identified important domains
magnitude of the economic burden of AF is increasing, mainly driven of AF care, and developed QIs for each domain. The development pro-
by AF-related complications and management costs, particularly cess involved conducting a systematic review of the literature, and using a
modified Delphi method25 to derive the final set of QIs and divide them
those associated with hospitalizations.2,12,13
into main and secondary QIs. The next step would be to conduct a feasi-
Data from the EURObservational Research Programme in AF
bility assessment of the developed QIs using existing AF registries.24
(EORP-AF) found that adherence to guideline-recommended thera-
Quality indicators may be divided into structural, process, and out-
pies in the treatment of AF is associated with lower mortality,14 yet come indicators.26 For each proposed QI, we provided relevant specifica-
large variability persists in the delivery of such therapies across tions, including numerator, denominator, measurement period, and
Europe.15,16 To improve the implementation of evidence-based medi- measurement duration. However, no care settings were suggested, be-
cine,17 some professional organizations have developed quality stand- cause the proposed QIs are applicable in both the inpatient and outpa-
ards, clinical indicators, and quality measures to evaluate and improve tient care. It is, thus, important to determine locally the clinical settingQuality indicators for the care and outcomes of adults with atrial fibrillation Page 3 of 17
during which QIs are applied in order to ensure the same processes of Excel spreadsheet. When available, the following information was
care are evaluated between healthcare providers. obtained for the extracted QIs: definition (including numerator, denomina-
tor, and exclusions), objective, type of QI (structural, process, outcome,
Members of the Working Group or PROM), domain of application, and potential data collection source.
The Working Group comprised members of the ECG Clinical Practice
Guidelines Task Force, as well as international experts in AF manage- Clinical practice guidelines and existing QIs
ment, patients with AF, and representatives from patient organizations. In addition to the systematic review outlined earlier, we reviewed relevant
Six domains of AF care were defined: (i) Patient assessment (baseline and
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
clinical practice guidelines and existing QIs from different professional
follow-up), (ii) Anticoagulation therapy, (iii) Rate control strategy, organizations (Table 1). The goal of the clinical practice guidelines review
(iv) Rhythm control strategy, (v) Risk factor management, and (vi) was to identify the recommendations with the strongest association with
Outcomes measures, including patient-reported outcome measures benefit or harm and to assess these recommendations against the ESC cri-
(PROMs). The names, affiliations, and conflicts of interest of the AF QIs teria for QIs (Table 2).24 Additionally, existing publications on QIs for
Working Group are provided in Supplementary material, Appendix 1. patients with AF were also reviewed and, when applicable, information
about the feasibility and/or validity of these measures was obtained.
Systematic review
Search strategy Data synthesis
We conducted a systematic review of the published literature in accor- Candidate QIs
dance with the Preferred Reporting Items for Systematic Review and A list of candidate QIs was derived from the aforementioned systematic
Meta-Analyses (PRISMA) statement27,28 (Supplementary material, review and classified into structural, process, or outcome measures
Appendix 2). We searched two online bibliographic databases: MEDLINE depending on the aspect of care being measured26. For each QI, a de-
R
and Embase via OVIDV. The initial search strategy was developed in tailed definition was provided in order to facilitate the evaluation process.
MEDLINE using keywords and, when available, Medical Subject Headings
(MesH) based on three main terms: ‘atrial fibrillation’, ‘quality indicators’,
and ‘outcome measures’ (Supplementary material, Appendix 3). The final Modified Delphi process
search strategies were then developed using an iterative process, which We used the modified Delphi process25, 29 to evaluate the candidate QIs
also included citations search, grey literature, and a hand search of the ref- and arrive at the final set of QIs. Instructions on the voting process, includ-
erence lists of the selected studies. ing QIs criteria (Table 2) were sent to the Working Group before the vote.
We included randomized controlled trials (RCTs) and observational All measures were independently graded by each member of the Group
studies, including local, national, and international registries. We excluded using the SurveyMonkey platform. Three rounds of voting were conducted,
systematic reviews, meta-analyses, editorial letters, and conference pro- with a teleconference after each round to discuss the results of the vote. In
ceedings. We only included the main publications of major trials and reg- the first voting round, we used a 9-point ordinal scale, where ratings of 1 to
istries from which our search obtained only their sub-studies. The search 3 signified that the QI was not valid; ratings of 4 to 6 meant that the QI was
was restricted to full-text articles published in the English language with a of uncertain validity; and ratings of 7 to 9 indicated that the QI was valid.
publication date between 1 January 2014 and 5 October 2019, to capture Candidate QIs were included if >_75% of the Working Group members
QIs and outcome measures for AF from contemporary practice. ranked them between 7 and 9, and were excluded if >_75% of the Working
Group members ranked them between 1 and 3. Indicators that did not fall
in the two categories above were carried forward to the second voting
Eligibility criteria round, where a 3-point scale (should not be included, maybe, and should
We included articles that fulfilled the following criteria: (i) the study popu- be included) was implemented, but the same percentage agreement (>_75%
lation was adult patients (>_18 years old) with AF, (ii) the study explicitly of the Working Group members) cut-off was used. The final round com-
stated at least one QI or outcome measure to define best practice for AF prised a binary, ‘yes’ or ‘no’ questionnaire to obtain the Working Group
diagnosis and/or management, (iii) the study provided specifications for members’ agreement on the proposed final set of QIs.
the QI or outcome measure (e.g. definition, data collection source,
method of reporting), (iv) RCT or registry, and (v) full-text publication.
No restrictions were applied to the presence of, or the type of, interven-
tion or comparison in the study.
Results
Search results
Study selection The literature search retrieved 2954 articles, of which 441 met the in-
A reference manager software (Zotero) was used for duplicates removal
clusion criteria (Figure 1). These articles were used to extract a total of
and data management. Two authors (Suleman Aktaa and Elena Arbelo) in-
352 candidate QIs (17 related to structure, 162 to process, and 173 re-
dependently examined the abstracts of the studies retrieved from the
search against the inclusion criteria. Disagreements were resolved through lated to outcomes) before the first voting round. Of these 34 QIs (19
discussion and review of the full text of the article when required. related to process and 15 related to outcomes) were selected by the
end of the second round (Table 3). Over 93% of the Working Group
members agreed on this final set of QIs in the third voting round.
Data extraction
The domains for AF care identified by the Working Group were:
The full texts of the included studies were independently reviewed by two
authors (Suleman Aktaa and Elena Arbelo). All QIs relevant to the agreed (i) Patient assessment (baseline and follow-up), (ii) Anticoagulation
six domains of AF care, namely: (i) Patient assessment (baseline and therapy, (iii) Rate control strategy, (iv) Rhythm control strategy, (v)
follow-up), (ii) Anticoagulation therapy, (iii) Rate control strategy, Risk factor management, and (vi) Outcome measures (including
(iv) Rhythm control strategy, (v) Risk factor management, and (vi) PROMs). For each domain, main, and for some secondary, QIs have
Outcomes measures (including PROMs) were extracted and listed on an been developed. Figure 2 shows the main QIs according to theirPage 4 of 17 E. Arbelo et al.
Table 1 Existing clinical practice guidelines and QIs used during the development process
Organization Type Year Country/Region
....................................................................................................................................................................................................................
ESC Guideline for the management of patients with atrial Clinical Practice Guidelines 2020 Europe
fibrillation23
ICHOM international standard set of outcome measures for QIs 2020 Worldwide
patients with atrial fibrillation134
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS Clinical Gractice Guidelines 2019 United States
Guideline for the management of patients with atrial fibrillation31
Canadian quality indicators for atrial fibrillation and atrial flutter20 QIs 2019 Canada
Harmonized outcome measures for use in atrial fibrillation patient QIs 2019 United States
registries and clinical practice154
ACC/AHA clinical performance and quality measures for adults QIs 2016 United States
with atrial fibrillation or atrial flutter18
ESC Guidelines for the management of atrial fibrillation Clinical Practice Guidelines 2016 Europe
developed2
NICE atrial fibrillation quality standard21 QIs 2015 United Kingdom
AHA/ACC/HRS Guidelines for the management of patients with Clinical Practice Guidelines 2014 United States
atrial fibrillation30
QI, quality indicators; AHA, American Heart Association; ACC, American College of Cardiology; ESC, European Society of Cardiology; ICHOM, International Consortium for
Health Outcomes Measurement; NICE, National Institute for Care and Health Excellence.
Table 2 Criteria for the development and evaluation of the ESC quality indicators for cardiovascular disease
Domain Criteria
....................................................................................................................................................................................................................
Importance QI reflects a clinical area that is of high importance (e.g. common, major cause for morbidity, mortality, and/or health-related
quality of life impairment).
QI relates to an area where there are diparities or suboptimal care.
QI implementation will result in an improvement in patient outcomes.
QI may address appropriateness of medical interventions.
....................................................................................................................................................................................................................
Evidence base QI is based on an acceptable evidence consistent with contemporary knowledge.
QI aligns with the respective ESC Clinical Practice Guideline recommendations.
....................................................................................................................................................................................................................
Specification QI has a clearly defined patient group to whom the measurement applies (denominator), including explicit exclusions.
QI has clearly defined accomplishment criteria (numerator).
....................................................................................................................................................................................................................
Validity QI is able to correctly assess what it is intended to, adequately distinguishes between good and poor quality of care, and com-
pliance with the indicator would confer health benefits.
....................................................................................................................................................................................................................
Reliability QI is reproducible even when data is extracted by different people, and estimates of performance on the basis of available data
are likely to be reliable and unbiased.
....................................................................................................................................................................................................................
Feasibility QI may be identified and implemented with reasonable cost and effort.
Data needed for the assessment is (or should be) readily available and easily extracted within an acceptable time frame.
....................................................................................................................................................................................................................
Interpretability QI is interpretable by healthcare providers, so that practitioners can understand the results of the assessment and take actions
accordingly.
....................................................................................................................................................................................................................
Actionability QI is influential to the current practice, where a large proportion of the determinants of adherence to the QI, are under the
control of healthcare providers.
This influence of QI on behaviour will likely improve care delivery.
QI is unlikely to cause negative unintended consequences.
QI, quality indicator.Quality indicators for the care and outcomes of adults with atrial fibrillation Page 5 of 17
Records idenfied through Addional records idenfied
Idenficaon
database searching through other sources
(n = 2954) (n = 9)
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
Records aer duplicates removed
(n = 2318)
Screening
Records screened Records excluded
(n = 2318) (n = 1162)*
Eligibility
Full-text arcles assessed Full-text arcles excluded
for eligibility (n = 766)**
(n = 1156)
Included
Studies included in
qualitave synthesis
(n = 441)
*Records excluded:
•Not populaon of interest (825): unrelated to cardiovascular disease (266), stroke (113), hearailure/congenital (81),
cardiovascular surgery (69), other arrhythmias (55), coronary artery disease (50), cardiovascular risk factors (35), valvular heart
disease (29), cardiovascular disease general (20), other (107)
•Study type (274): abstract (201), study protocol (22), non-systemac review (20), systemac-review/meta-analysis (7),
leer/note/survey (11), cost-effecveness analysis (5), editorial (4), basic research (4).
•Sub-studies of main publicaons (53)
•Other (10)
**Full text arcles excluded due to the lack of clear definions and/or specificaons of the quality indicators and/or outcome measures.
Figure 1 PRISMA flow diagram for selection of included studies.
respective domain of care. The full set of main and secondary QIs,
01MQI1: Proportion of patients with cardio-embolic risk as-
alongside their definitions, proposed measurement period (the time
sessment using CHA2DS2-VASc score
point at which the assessment is performed), measurement duration
Numerator: Number of patients with AF who have their CHA2DS2-
(the time frame needed for enough cases to be collected), and when
VASc score documented at the time of diagnosis and at every follow-up
applicable, the corresponding ESC Clinical Practice Guidelines rec-
appointment.
ommendations are illustrated in APPENDIX 4. For each QI, a unique
Denominator: Number of patients with AF.
code was developed using the domain number and indicating
01MQI2: Proportion of patients with bleeding risk assessment
whether the QI is main or secondary.
using a validated method, such as the HAS-BLED score
Numerator: Number of patients with AF who have their bleeding risk
Quality indicators assessment documented at the time of diagnosis and at every follow-up
Domain 1: Patient assessment (baseline and follow-up) appointment using a validated bleeding risk score.
Stroke prevention is the cornerstone of the AF patient management Denominator: Number of patients with AF.
pathway, and ‘avoid stroke/anticoagulation’ is the ‘A’ of the ABC 01MQI3: Proportion of patients with a measurement of their
pathway32, within the 2020 ESC guidelines23. serum creatinine (or creatinine clearance)
Stroke risk in AF is not homogeneous and depends on the pres- Numerator: Number of patients with AF who have their serum creatinine
ence of various stroke risk factors33. The CHA2DS2-VASc score is checked at the time of diagnosis and at every follow-up appointment.
recommended to assess stroke risk where the default should be to Denominator: Number of patients with AF.
offer stroke prevention, unless the patient is low risk; hence use thePage 6 of 17 E. Arbelo et al.
Table 3 Primary (green) and secondary (yellow) quality indicators for AF diagnosis and management
Code Quality indicators
....................................................................................................................................................................................................................
Domain 01: Patient assessment (at baseline and follow-up)
01MQI1 Proportion of patients with cardio-embolic risk assessment using CHA2DS2-VASc score
01MQI2 Proportion of patients with bleeding risk assessment using a validated method, such as the HAS-BLED score
01MQI3 Proportion of patients with a measurement of their serum creatinine (or creatinine clearance)
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
01SQI1 Proportion of people >_65 years of age with risk factors for AF who have pulse check
01SQI2 Proportion of patients with AHREs detected on implantable cardiac devices who undergo further cardiovascular evaluation
01SQI3 Proportion of cryptogenic stroke patients who have been screened for AF
01SQI4 Proportion of patients with an ECG documentation of AF
01SQI5 Proportion of patients who have been engaged in shared decision making when deciding treatment strategy
Domain 02: Anticoagulation
02MQI1 Proportion of patients who are appropriately prescribed anticoagulation according to CHA2DS2-VASc score*
02MQI2 Proportion of patients with a CHA2DS2-VASc score of 0 for men and 1 for women who are inappropriately prescribed long-term
anticoagulation
02MQI3 Proportion of patients with ‘appropriate anticoagulation’ at every follow-up visit, defined as:
c. TTR**>_70% for vitamin-K antagonist.
d. Appropriate dose for NOAC according to manufacturer recommendations.
Domain 03: Rate control
03MQI1 Proportion of patients with permanent AF (i.e. where no attempt to restore sinus rhythm is planned), who are inappropriately
prescribed antiarrhythmic drugs
03SQI1 Proportion of patients with LVEF_75 years, Diabetes mellitus, Stroke, Vascular disease, Age 65–74 years,
Sex category (female); ECG, electrocardiogram; HAS-BLED, Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile INR, Elderly (>65 years),
Drugs/alcohol concomitantly; LVEF, left ventricular ejection fraction; NOAC, non-vitamin K antagonist oral anticoagulants; PVs, pulmonary veins; TTR, time in therapeutic range.
*Appropriateness of anticoagulation prescription is defined as CHA2DS2-VASc score of >_1 for men and >_2 for women in the 2020 ESC Guidelines23. The 2014 ACC/AHA
Guidelines (and 2019 focused update) define anticoagulation prescription appropriateness and CHA2DS2-VASc score of >_2 for men and >_3 for women30,31.
**TTR calculated using Rosendaal method.
***Crude and risk-adjusted rates (risk adjustment should, as a minimum, consider age, sex, and comorbidities).Quality indicators for the care and outcomes of adults with atrial fibrillation Page 7 of 17
1. Patient assessment
CHA2DS2-VASc
Bleeding risk
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
Serum creatnine
5. Risk factor
management 2. Anticoagulation
Modifiable risk factor OAC prescribed for high
identification, including CHA2DS2-VASc
blood pressure, obesity, Inappropriate OAC for
obstructive sleep 6. Outcome measures low CHA2DS2-VASc
apnoea, alcohol excess,
All cause morality TTR ≥70% / appropriate
lack of exercise, poor
Ischaemic stroke / TIA NOAC dose
glycaemic control and
smoking Life-threatening / major
bleeding
Procedure-related death
Procedure-/drug-related
serious adverse events
HRQoL
4. Rhythm control
Inappropriate use of
class IC AAD in
structural heart disease
Inappropriate use of 3. Rate control
dofetilide/sotalol in Inappropriate AAD in
ESRD or dialysis permanent AF
CA for symptomatic
paroxysmal/persistent
AF after one class I or
class III AAD
Figure 2 Domains of AF care with their respective main quality indicators. AAD, antiarrhythmic drug; AF, atrial fibrillation; CA, catheter ablation;
ESRD, end-stage renal disease; HRQoL, health-related quality of life; NOAC, non-vitamin K antagonist oral anticoagulant; OAC, oral anticoagulants;
TTR, time in therapeutic range; TIA, transient ischaemic attack.
CHA2DS2-VASc score to initially define low risk patients (CHA2DS2- Based on a Patient-Centered Outcomes Research Institute (PCORI)
VASc score 0 in males, 1 in females) who do not need antithrombotic systematic review and evidence appraisal, the best validated bleeding
therapy (indicator 01MQI1). The subsequent step is to offer stroke risk score is the HAS-BLED score36 . While stroke and bleeding risks
prevention in those with 1 or more risk factors (CHA2DS2-VASc track each other, the evidence shows that a formal bleeding risk
score >_1 in males, >_2 in females). Since stroke risk is dynamic, and score (HAS-BLED) is superior to stroke risk scores (e.g. CHADS2,
influenced by ageing and incident risk factors, risk reassessment CHA2DS2-VASc) for assessing bleeding risk37,38. A strategy for dy-
should occur at every follow-up visit34. namic bleeding risk assessment using the HAS-BLED score has been
Bleeding risk changes over time as well and should also be assessed shown to reduce bleeding risk and to increase oral anticoagulation
at every patient contact, initially to identify modifiable bleeding risks (OAC) use39.
that should be mitigated, and to identify the ‘high bleeding risk’ patient Given that renal function has implications for both stroke and bleeding
who should be scheduled for early follow-up35 (indicator 01MQI2). risk40, as well as prescriptions of OAC (choice of agent and dose),Page 8 of 17 E. Arbelo et al.
However, screening for AF should be accompanied by confirming
01SQI1: Proportion of people 65 years of age with risk factors
the diagnosis by traditional means, such as by 12-lead ECG or >30 s
for AF who have pulse check
recording of a single-lead ECG, Holter monitor, or event recorder
Numerator: Number of people >_65 years of age with risk factors for AF
(indicator 01SQI4). Following the diagnosis, a dialogue between
who have a documentation of pulse check (or ECG) to identify rhythm.
treating physician and patient to ensure patient involvement in deci-
Denominator: Number of people >_65 years of age with risk factors for
sion making is recommended23,54 . Thus, the indicator 01SQI5 cap-
AF.
tures shared decision making when deciding on the treatment
01SQI2: Proportion of patients with atrial high-rate episodes
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
strategy.
(AHREs) detected on implantable cardiac devices who undergo
further cardiovascular evaluation
Numerator: Number of patients with AHREs detected on implantable Domain 2: Anticoagulation
cardiac devices who have documentation of complete cardiovascular Oral anticoagulation is an essential part of AF management, and the
evaluation. ESC 2020 Guidelines recommend oral anticoagulation for stroke pre-
Denominator: Number of patients with atrial high-rate episodes vention in males with CHA2DS2-VASc scores of >_1, and in females
detected on implantable cardiac devices. with scores of >_223. Accordingly, it is important that a set of QIs to
01SQI3: Proportion of cryptogenic stroke patients who have regularly assess the proportion of patients with CHA2DS2-VASc score
been screened for AF >_1 in males, >_2 in females who are offered stroke prevention (indica-
Numerator: Number of patients with cryptogenic stroke* who have tor 02MQI1), as well as the inappropriate use of long-term antithrom-
documentation of AF screening using continuous ECG recording. botic therapy in low risk patients (CHA2DS2-VASc score 0 in males,
Denominator: Number of patients with cryptogenic stroke with no pre- and 1 in females) (indicator 02MQI2).
vious history of AF. Assessment of the quality of anticoagulation is also important. If
01SQI4: Proportion of patients with an ECG documentation of patients are taking a non-vitamin K antagonist oral anticoagulant
AF (NOAC), the label-adherent dose of the respective NOAC should
Numerator: Number of AF patients with documentation of an ECG
confirming AF diagnosis. 02MQI1: Proportion of patients who are appropriately pre-
Denominator: Number of AF patients. scribed anticoagulation according to CHA2DS2-VASc score**
01SQI5: Proportion of patients who have been engaged in Numerator: Number of patients with AF who have CHA2DS2-VASc
shared decision making when deciding treatment strategy score of >_1 for men and >_2 for women and are prescribed anticoagula-
Numerator: Number of AF patients with a documentation of patient en- tion for AF.**
gagement when deciding treatment strategy. Denominator: Number of patients with AF who have CHA2DS2-VASc
Denominator: Number of AF patients. score of >_1 for men and >_2 for women and are eligible for anticoagula-
tion, with no contraindication or refusal.**
02MQI2: Proportion of patients with a CHA2DS2-VASc score of
0 for men and 1 for women who are inappropriately prescribed
regular measurements of serum creatinine or creatinine clearance (based long-term anticoagulation
on the Cockcroft-Gault formula) are needed, the frequency of which is Numerator: Number of patients with AF who have CHA2DS2-VASc
determined by the renal function at baseline41 (indicator 01MQI3). score of 0 for men and 1 for women and are inappropriately prescribed
Asymptomatic AF is associated with a higher risk of stroke and mor- long-term anticoagulation for AF.
tality compared with symptomatic AF. An observational study indi- Denominator: Number of patients with AF who have CHA2DS2-VASc
cated that the application of standard care treatments for subclinical score of 0 for men and 1 for women and do not have other indication for
AF detected on screening improves outcomes45,, and a systematic re- anticoagulation.
view and economic analysis suggested that screening programmes for 02MQI3: Proportion of patients with ‘appropriate anticoagula-
AF are likely to represent a cost-effective use of resources46,. Thus, tion’ at every follow-up visit, defined as:
screening for AF amongst people >_65 years of age by checking their a. Time in therapeutic range TTR70% for vitamin-K
pulse may have therapeutic implications as these individuals need to be antagonist.
considered for thromboprophylaxis (indicator 01SQI1). b. Appropriate dose for NOAC according to manufacturer
To that end, atrial high-rate episodes (AHRE) detected by recommendations.***
implanted cardiac devices, which may represent asymptomatic AF, Numerator: Number of patients with AF who have appropriate anticoa-
should be investigated47,48. Ideally, AHRE detection should be gulation defined as TTR>_70% for vitamin-K antagonist, and appropriate
performed at every device interrogation, including home monitor- dose for NOAC according to manufacturer recommendations.***
ing transmission as it determines whether or not subclinical AF is Denominator: Number of patients with AF on anticoagulation.
confirmed and whether anticoagulation and/or regular follow-up
is warranted23 (indicator 01SQI2). Furthermore, the detection of **Appropriateness of anticoagulation prescription is defined as CHA2DS2-VASc
score of >_1 for men and >_2 for women in the 2020 ESC Guidelines. The 2014
previously unknown AF following a stroke has relevant implica- ACC/AHA Guidelines (and 2019 focused update) define anticoagulation pre-
tions for secondary prevention49,50. Thus, it is recommended to scription appropriateness and CHA2DS2-VASc score of >_2 for men and >_3 for
screen for AF following a cryptogenic stroke 2352–54 (indicator women.23,30,31
***Manufacturer recommendations are defined in APPENDIX 5.
01SQI3).Quality indicators for the care and outcomes of adults with atrial fibrillation Page 9 of 17
be prescribed and the proportion appropriately dosed is indicative of
04MQI1: Proportion of patients with structural heart disease
quality of care. Regular audits should be performed to ensure that un-
who are inappropriately prescribed class IC antiarrhythmic
der- or over-dosing of the respective NOAC does not occur, given
drugs
the association with worse outcomes55–57 (indicator 02MQI3). Oral
Numerator: Number of patients with AF who have structural heart dis-
anticoagulation can also be offered as a well-managed vitamin K an-
ease and are inappropriately prescribed class IC antiarrhythmic drugs.
tagonist (VKA) (e.g. warfarin, acenocoumarol, phenprocoumon, etc.),
Denominator: Number of patients with AF who have structural heart
with a high (>_70%) time in therapeutic range (TTR) calculated using
disease.
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
the Rosendaal method, with international normalized ratio (INR)
04MQI2: Proportion of patients with end-stage kidney disease
2.0–3.0. High TTR has been associated with low rates of stroke and
who are inappropriately prescribed dofetilide or sotalol
bleeding, as well as reduced mortality58–60. Thus, the proportion of
Numerator: Number of patients with AF who have end-stage kidney disease
patients with TTR >_ 70% is a good QI of anticoagulation control for
and/or on dialysis$$ and are inappropriately prescribed dofetilide or sotalol.
patients on VKA.
Denominator: Number of patients with AF who have end-stage kidney
disease, including patients on dialysis.
04MQI3: Proportion of patients with symptomatic paroxysmal
or persistent AF who are offered AF catheter ablation after
Domain 3: Rate control
failure of, or intolerance to, one class I or class III antiarrhyth-
Rate control is an integral part of AF management, and may be suffi-
mic drug
cient to improve AF-related symptoms61. In patients for whom a de-
Numerator: Number of patients with paroxysmal or persistent AF who
cision has been made not to restore or maintain sinus rhythm
are offered catheter ablation after the failure of, or intolerance to, one
(permanent AF), rate control can be achieved by rate-limiting medi-
class I or class III antiarrhythmic drug.
cations (e.g. beta-blockers, digoxin, diltiazem, or verapamil). The use
Denominator: Number of patients with paroxysmal or persistent AF with no
of antiarrhythmic drugs, such as amiodarone, dronedarone, or sotalol
contraindications (or refusal) to catheter ablation who remain symptomatic on,
for rate control is not recommended when no attempt to restore si-
or intolerant to, one class I or class III antiarrhythmic drug.
nus rhythm is planned62–65 (indicator 03MQI1).
The use of certain types of rate control drugs, such as non-
dihydropyridine calcium-channel blockers can influence outcomes in
patients with heart failure and/or left ventricular ejection fraction
(LVEF) ofPage 10 of 17 E. Arbelo et al.
A QI to assess the complete electrical isolation (entrance and exit
06.1MQI1: Annual rate of all-cause mortality*
block) of the pulmonary veins (PVs) during AF catheter ablation pro-
Numerator: Number of patients with AF who died during the measure-
cedures (indicator 04SQI1) was developed given that this is the de-
ment duration.
sired outcome of AF ablation69,73,74,88–99. In addition, the indicator
Denominator: Number of patients with AF.
04SQI2 assesses the consideration of cardioversion for patients with
06.1MQI2: Annual rate of ischaemic stroke or transient ischae-
new-onset persistent AF.
mic attack*
Numerator: Number of patients with AF who had documented ischae-
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
mic stroke or transient ischaemic attack during the measurement
Domain 5: Risk factor management
duration.
The Working Group considered the role of risk factors in AF and
Denominator: Number of patients with AF.
developed a QI accordingly (indicator 05MQI1). Recent research
has highlighted the potential benefits of risk factor management as *Crude and risk-adjusted rates (risk adjustment should, as a minimum, consider
upstream non-invasive therapy to lower the risk of AF progres- age, sex, and comorbidities.
sion and recurrence100–106. A large proportion of these risk fac-
tors are lifestyle related and, therefore, are amenable to be
targeted and modified107. It is recommended that in the assess-
ment of AF patients, practitioners actively evaluate and document 06.1SQI1: Annual rate of cardiovascular mortality*
these modifiable risk factors, such as smoking, obesity100,102,108, Numerator: Number of patients with AF who died from cardiovascular
physical inactivity109–111, alcohol intake105,112–114, sleep115 ap- cause during the measurement duration.
noea116,117, hypertension115,118,119, and poor glycaemic con- Denominator: Number of patients with AF.
trol120. Where necessary, appropriate education, support, and 06.1SQI2: Annual rate of cardiovascular hospitalization*
intervention (e.g. smoking cessation options, continuous positive Numerator: Number of patients with AF who had unplanned hospitaliza-
airway pressure (CPAP), exercise prescription, etc.) can be pro- tion for a cardiovascular cause during the measurement duration.
vided to the patient to address the risk factors that may improve Denominator: Number of patients with AF.
health outcomes. 06.1SQI3: Annual rate of overall thromboembolic events*
Numerator: Number of documented AF-related thrombo-embolic
events during the measurement duration.
05MQI1: Proportion of patients who have their modifiable risk
Denominator: Number of patients with AF.
factors identified
06.1SQI4: Annual rate of clinician-reported symptom status
Numerator: Number of patients with AF who have their modifiable risk
assessment
factors (e.g. blood pressure, obesity, obstructive sleep apnoea, alcohol
Numerator: Number of patients with AF who had their clinician-
excess, lack of exercise, poor glycaemic control, and smoking) identified.
reported symptom status assessed using a validated tool (e.g. EHRA
Denominator: Number of patients with AF.
symptom score) during the measurement duration.
Denominator: Number of patients with AF.
*Crude and risk-adjusted rates (risk adjustment should, as a minimum, consider
Domain 6: Outcome measures age, sex, and comorbidities.
Consequences of the disease
Reducing the risk of death is one of the primary aims of AF man-
agement, and healthcare in general23. As such, annual assessment of
crude and risk-adjusted rates of all-cause mortality is recommended
AF-related symptoms can be a useful subjective measure of both the
(indicator 06.1MQI1). Risk adjustment should, as a minimum, con-
clinical consequences of AF and the success of rate- and rhythm con-
sider age, sex, and comorbidities. In addition, the inclusion of lifestyle
trol treatment from the patients’ perspective. Using a validated
factors (e.g. smoking status, body mass index, physical activity, and al-
method, such as the modified European Heart Rhythm Association
cohol intake) provides a better insight to the adjustment process.
(EHRA) score121 is recommended to assess symptom status (indica-
Given that ischaemic stroke is a major complication of AF and, that
tor 06.1SQI4).
most AF patients (CHA2DS2-VASc score of >_1 in men and >_2 in
women) will be eligible for stroke prevention, the overall and risk-
adjusted annual incidence of stroke and, separately, transient ischae-
mic attack should be recorded as a QI (indicator 06.1MQI2). Other Complications of treatment
outcomes measures, which may provide an illustration of the quality OAC treatment conveys an increased risk of major bleeding.
of AF care, include the rate of cardiovascular mortality (indicator However, bleeding complications can also occur in the absence of
06.1SQI1), cardiovascular hospitalization (indicator 06.1SQI2), over- OAC treatment122. The incidence of life-threatening or major bleed-
all thrombo-embolic events (indicator 06.1SQI3), and clinician- ing events, defined by the International Society of Thrombosis and
reported AF symptom status (indicator 06.1SQI4). Haemostasis criteria123,124, should be reported annually as a QI (indi-
In the ABC pathway of AF management mentioned earlier, the ‘B’ cator 06.2MQI1). The annual rate of haemorrhagic stroke is of partic-
component pertains to ‘better’ symptom management33. Many AF ular importance (indicator 06.2SQI1) and should be documented as
patients may not be overtly symptomatic. However, assessment of a QI.Quality indicators for the care and outcomes of adults with atrial fibrillation Page 11 of 17
06.2MQI1: Annual rate of life-threatening or major bleeding 06.3MQI1: Proportion of patients with health-related quality of
events& life assessment
Numerator: Number of patients with AF on anticoagulation who had Numerator: Number of patients with AF who have their health-related
documented life-threatening or major bleeding events during the mea- quality of life assessed at the time of diagnosis and least annually after-
surement duration. wards using a validated instrument.
Denominator: Number of patients with AF on anticoagulation. Denominator: Number of patients with AF.
06.2MQI2: Annual rate of procedure-related&& 30-day 06.3SQI1: Proportion of patients with patient-reported symp-
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
mortality tom status assessment
Numerator: Number of patients with AF who died due to an invasive Numerator: Number of patients with AF who have their patient-
procedure for AF management during the measurement duration. reported symptom status assessed at the time of diagnosis and least an-
Denominator: Number of patients with AF treated with invasive nually afterwards using a validated instrument.
procedures. Denominator: Number of patients with AF.
06.2MQI3: Annual rate of procedure-related&& major compli- 06.3SQI2: Proportion of patients with physical function
cations or drug-related serious adverse events$ assessment
Numerator: Number of patients with AF who had documented major Numerator: Number of patients with AF who have their physical func-
procedural complications and/or drug-related serious adverse events tion assessed at the time of diagnosis and at every follow-up appointment
during the measurement duration. using a validated instrument.
Denominator: Number of patients with AF. Denominator: Number of patients with AF.
06.2SQI1: Annual rate of haemorrhagic stroke 06.3SQI3: Proportion of patients with emotional well-being (in-
Numerator: Number of patients with AF who had documented haemor- cluding anxiety and depression) assessment
rhagic stroke during the measurement duration. Numerator: Number of patients with AF who have their emotional
Denominator: Number of patients with AF on anticoagulation. well-being (including anxiety and depression) assessed at the time of diag-
nosis and at every follow-up appointment using a validated instrument.
Denominator: Number of patients with AF.
06.3SQI4: Proportion of patients with cognitive function
assessment
AF procedure-related deaths occurring within the first 30 days fol- Numerator: Number of patients with AF who have their cognitive func-
lowing catheter-based ablation, surgical ablation procedure, hybrid tion assessed at the time of diagnosis and at least annually afterwards us-
catheter and surgical ablation, left atrial appendage closure/occlusion ing a validated instrument.
(device), left atrial appendage ligation/excision (surgical), electrical Denominator: Number of patients with AF.
cardioversion, or pacemaker implantation, should be reported annu-
ally as a QI (indicator 06.2MQI2). Furthermore, any procedure-
related major complication or drug-related serious adverse event,
defined as any untoward medical occurrence that results in death, life (AFEQT) or the Atrial Fibrillation Severity Scale (AFSS)]131–134.
life-threatening outcomes, hospitalization (initial inpatient hospitaliza- Both the SF-12 and the AFEQT are validated, psychometrically ro-
tion or prolongation of existing hospitalization for >_24 h), or perma- bust assessments of HRQoL, and are recommended by the
nent injury, should be reported in real-time according to local or International Consortium of Healthcare Outcome Measures
national policy, and annually as a marker of quality (indicator (ICHOM) for AF135. Regardless of which validated tool is employed,
06.2MQI3). Although a single QI is suggested for procedural compli- it is important that the same PROM is used consecutively to assess
cations (e.g. atrio-oesophageal fistula, cardiac tamponade, PV steno- HRQoL to permit temporal comparison of scores and allow the de-
sis, phrenic nerve palsy, etc.), and drug-related adverse events (e.g. termination of response to treatment.
arrhythmias, sudden cardiac death, etc.), individual events may be col- Determining the impact of AF and its treatment on the patient are
lected in each centre for local monitoring and between-centre important considerations in the management of AF and may contrib-
comparisons. ute to patient and healthcare provider decisions regarding continua-
tion/cessation of certain treatments and/or initiating alternatives. In
addition to HRQoL, the assessment of other PROMs, such as
Patient-reported outcomes patient-reported symptom status (indicator 06.3SQI1), physical func-
PROMs are important determinants of the patients’ perceived tioning (indicator 06.3SQI2), emotional well-being (indicator
quality and success of treatment125–127. The 2020 ESC Guidelines 06.3SQI3), and cognitive function (indicator 06.3SQI4), could also be
recommend that patient-reported outcomes should be routinely considered. The assessment of HRQoL, patient-reported symptom
collected to measure treatment success and improve patient status, physical functioning, and emotional well-being is recom-
care23. Health-related quality of life (HRQoL) is considered the mended at baseline and once to twice annually, while the assessment
main QI and should be assessed at baseline and at follow-up visits of cognitive function is recommended at baseline and annually there-
(indicator 06.3MQI1). after, given that it may show little variation over a shorter period of
Several validated tools are available to measure general HRQoL128 time. Validated tools, such as those recommended by the ICHOM
[e.g. the Short-Form 12 (SF-12)]129, while others specifically measure for AF135 (PROMIS Global Health for physical and emotional well-
AF-specific HRQoL130 [e.g. the Atrial Fibrillation Effect on QualiTy of being, and PROMIS for cognitive function) can be used.Page 12 of 17 E. Arbelo et al.
Comparison with other quality metrics behaviour and practice, and were not included in the final set of indi-
Table 4 shows a comparison between the 2020 ESC QIs for AF and cators. Other QIs, such as the reintroduction of OAC after a severe
quality metrics from other professional organizations, such as the bleeding event, once the condition leading to the bleeding event has
American College of Cardiology and the American Heart been appropriately addressed59,157, and the use of strict versus le-
Association (ACC/AHA), the National Institute for Health and Care nient rate control treatment158, were proposed in the initial set of
Excellence (NICE), the Canadian Cardiovascular Society (CCS), and candidate QIs, but were deemed difficult to operationalize, and, thus,
ICHOM. There are major differences between the process QIs pro- were not included.
Downloaded from https://academic.oup.com/europace/advance-article/doi/10.1093/europace/euaa253/5898806 by guest on 10 September 2020
posed here, and those developed by ACC/AHA, NICE, and CCS. Conversely, and to emphasize that improving outcomes is the ulti-
These differences may be explained by the variation in clinical prac- mate aim of a quality of care assessment (Figure 1), particular atten-
tice guidelines endorsed by different societies and/or local needs to tion was given to outcome QIs. The term ‘outcome measures’ was
address certain gaps in AF care. Outcome QIs were relatively similar used separately and in different variations in the systematic review
compared to those proposed by ICHOM. search strategy (APPENDIX 3). The outcome QIs selected are appli-
cable to all domains of AF care, and are in line with the recent
ICHOM recommendations159.
Discussion One important type of outcome QIs is PROMs, which are increas-
ingly used in everyday practice. Although a structured methodology
Evaluating the quality of care delivered and measuring meaningful for developing and reporting PROMs exists159, there is uncertainty
outcomes of both the condition and its treatment have become an around the best instruments to collect such measures. By defining
essential element of modern healthcare136. AF is the most common specific PROMs and recommending tools for their measurement, the
cardiac arrhythmia, affecting 2–4% of the population, and is a major Working Group hopes to promote PROMs use in a systematic man-
cause of significant morbidity137. Although evidence suggests that ad- ner. However, developing outcome QIs to measure the results of
herence to guideline-recommended therapies for AF is associated PROMs assessment, as well as their temporal trends may not be fea-
with improved outcomes138,139, data from AF registries continue to sible in contemporary practice. Thus, process QIs to measure and en-
show room for improvement and significant geographical variation in courage PROMs assessment were developed instead.
AF quality of care and outcomes57,58,140–153. QIs have been devel- The Working Group acknowledges that high-quality evidence sup-
oped to evaluate the quality of AF care18,20,21,154,155. Furthermore, porting PROMs use is limited, widely accepted tools to collect them
QIs provide the mechanism to assess the effectiveness of quality im- are lacking, and little experience exists on how PROMs can guide AF
provement initiatives156. However, standardized measures to facili- treatment decisions. The same argument can be levelled at shared
tate ongoing efforts to quantify the adherence to guidelines are decision making in AF management. However, these aspects of AF
needed. care were deemed essential by the Working Group, thus QIs for
The present document is the first effort undertaken by the ESC to PROMs and shared decision making were developed.
develop a set of QIs to assess the quality of care for patients with AF. The patient’s perspective is a fundamental element of optimal AF
Using the ESC methodology for QIs development24, we have estab- care given that most therapies are aimed at improving patients’ symp-
lished a comprehensive set of QIs for AF care, which are supported toms, well-being, and overall QoL. Measuring patient-centred out-
by evidence and underpinned by expert consensus. Thus, they pro- comes in a standardized way may allow comparison of performance,
vide tools to quantify the quality of AF care and can be used as a basis enable clinicians to learn from each other, and improve the care we
for quality improvement. The simultaneous development of the ESC provide to our patients. However, further validation of the tools and
AF QIs and the ESC Clinical Practice Guidelines for AF facilitated methods used to collect the patient’s perspective in routine clinical
seamless incorporation of QIs within the guidelines document. As practice is needed. As such, these tools may be used to guide the de-
such, a summary form of the developed QIs is embedded within the velopment of, and the effect of, treatment strategies for AF patients.
ESC Clinical Practice Guidelines for AF, with the hope of enhancing The methodology used for the selection of QIs has limitations. We
their dissemination and, therefore, uptake into clinical practice23. relied on expert opinion to arrive at the final set of QIs following the
This document is the result of an international collaboration (12 comprehensive systematic review of the literature. A different panel
countries) from seven professional societies/associations with a of experts may have selected different QIs. We addressed this chal-
Working Group consisting of a wide range of stakeholders, including lenge by using the modified Delphi method, and by involving AF spe-
patients. In addition, the application of ESC criteria ensured that de- cialists with different areas of expertise, as well as patients and
veloped QIs are not only based on evidence, but also cover broad representatives from AF patient associations.
aspects of AF care where there is a gap in care delivery, potential for Another challenge is that, if considered in isolation, QIs may cause
quality improvement, and the availability of reliable data collection some unintended consequences, such as anticoagulation prescription
sources. To that end, different types of QIs including structural, pro- for patients with very high bleeding risk or recommending catheter
cess, and outcome indicators26 were included in the initial set of can- ablation for frail patients with major risk factors for AF recurrence.
didate QIs. We have sought to circumvent this issue by clearly defining eligible
The Working Group, however, considered structural QIs, such as patients for each QI and specifying relevant exclusions. The suggested
the volume of catheter ablation cases for centres and individual oper- QIs are intended to drive holistic patient assessments and tailor treat-
ators not to be directly under the control of healthcare providers. ments to individual patients to improve patient care. More refine-
Thus, structural QIs, although important, were given less priority ment of these QIs and/or their definitions may be needed in the
compared with other process QIs that may influence providers’ future when more ‘real-world’ and feasibility data become available.You can also read
